交岔路口事故通常涉及雙方(雙車、車與行人)。我國之機車路權與行人環境長期遭到漠視，使弱勢使用者較易涉入嚴重事故，然現有改善措施成效不彰，是以透過分析釐清事故嚴重度之影響因素，並對此作出改善。涉入事故的雙方可能會遭受完全不同程度的傷害，要正確辨認雙方事故的真正因果關係，有必要同時評估雙方事故的嚴重度。在事故嚴重度研究中，已使用各種潛在類別依序模型來捕捉事故發生的異質性。然而，多數研究為單變量，並不適用雙車事故。本研究提出一種潛在類別參數化相關性雙變量依序普羅比模型，以研究交岔路口的雙方事故。
本研究以2018至2020年間32,308件臺北市路口雙方事故作為分析對象，將嚴重度分為財損、輕傷/可能受傷、致命/明顯受傷三個等級。透過潛在類別模式確定兩群為最佳群數，是為低風險群與高風險群。本模型不僅可參數化雙方事故嚴重度的門檻值和事故相關性，還可根據特徵將雙方事故分類為不同的風險群，從而更好地理解雙方事故下的變數。普通事故群(OCS)主要涉及雙車碰撞之機車事故；高嚴重事故群(HCS)由行人與機車騎士等弱勢使用者組成，推測主要出現在車流量大的地區。
基於估計結果指出一些潛在因素，例如駕駛(老年人)、違規行為(安全裝備、讓車或肇逃)和運具類型(四輪車輛、二輪車輛或行人)，交通工程三要素中，人、車、路皆存在為影響嚴重度之風險因素。透過彈性效應，OCS群於致命/明顯受傷之彈性值高於HCS群，變數型態則以運具類型的致命/明顯受傷值最高，強調其影響性。本研究藉此希望減少路口違規行為，並預防大型車輛事故。
結果說明特定某方因素對嚴重度的影響大於雙方通用因素，並對路口事故提供寶貴的見解，透過結合傳統交通3E(工程、教育、執法)與鼓勵成為第4E，制定相應的安全措施，以降低未來事故發生件數與嚴重度，建議相關單位執行本研究提出之策略，並提升全民之行車觀念。本研究最後透過分析事故嚴重度與肇責釐清事故因果關係，可使保險相關風險得到管控。

Street intersection crashes often involve two parties (vehicle-vehicle and vehicle-pedestrian). The disregard for the right-of-way of motorcycles and the pedestrian environment in our country has been ignored, making vulnerable users more prone to serious accidents. However, existing improvement has been proven ineffective. Therefore, it is necessary to analyze the factors affecting the injury severity and to make improvements accordingly. The parties involved in crashes can vary considerably. To accurately identify the causality of a two-party crash, it is necessary to assess the damage of both parties simultaneously. While the latent class ordinal model has been used in crash severity studies to capture heterogeneity in crash propensity, most are univariate. They are inappropriate for the context of two-vehicle crashes. We propose a latent class parameterized correlation bivariate generalized ordered probit (LCp-BGOP) model to examine two-party crashes at intersections in the study.
This study collected 32,308 cases of two-party crashes at street intersections in Taipei City from 2018 to 2020. Injury severity is categorized into three levels: property damage only, minor/possible injury, and fatal/evident injury. Here are two classes, low-risk and high-risk, determined as the optimal class number through the latent class method. The LCp-BGOP parameterizes the thresholds and within-crash correlations of two-party crash severity, and it classifies the crashes into distinct risk groups based on risk variables, thereby better understanding variables in intersection crashes. According to our model, the Ordinary Crash Severity (OCS) group mainly involves two-vehicle crashes colliding with motorcycles; the High Crash Severity (HCS) group comprises vulnerable road users like pedestrians and cyclists, mainly in mixed traffic with high volumes.
Our model-based estimation points out several potential factors, such as drivers (elderly), violations (safety equipment, yielding to vehicles, or hit-and-run), and modes (four-wheeled vehicles, two-wheeled vehicles, or pedestrians). Three elements of traffic engineering, namely people, vehicles, and roads, are some existing risk factors that can influence severity. Through the elasticity effects, the OCS group has a higher magnitude of fatal/evident injury than the HCS does. By variable patterns, the mode of mobility exhibits the highest fatal/evident injury values, underscoring its significant influence. Accordingly, we hope to reduce violations at intersections and prevent large vehicle crashes.
The results show that the party-specific factors contribute to injury severity more than generic factors do, providing invaluable insight into intersection crashes from the perspective of reducing two-party collisions. By integrating the traditional traffic 3E (Engineering, Education, and Enforcement) with Encouragement into 4E, we develop the corresponding safety measures to reduce the frequency and severity of future crashes. It is recommended that authorities implement the strategies proposed in this study and enhance public awareness of driving. Finally, this study clarifies causal relationships in accidents by analyzing crash severity and fault determination, enabling risk management for insurance.

Balusu, S. K., Pinjari, A. R., Mannering, F. L., & Eluru, N. (2018). Non-decreasing threshold variances in mixed generalized ordered response models: A negative correlations approach to variance reduction. Analytic methods in accident research, 20, 46-67.
Benlagha, N., & Charfeddine, L. (2020). Risk factors of road accident severity and the development of a new system for prevention: New insights from China. Accident Analysis & Prevention, 136, 105411.
Bhat, C. R. (1997). An endogenous segmentation mode choice model with an application to intercity travel. Transportation science, 31(1), 34-48.
Blows, S., Ivers, R. Q., Connor, J., Ameratunga, S., & Norton, R. (2003). Car insurance and the risk of car crash injury. Accident Analysis & Prevention, 35(6), 987-990.
Cerwick, D. M., Gkritza, K., Shaheed, M. S., & Hans, Z. (2014). A comparison of the mixed logit and latent class methods for crash severity analysis. Analytic methods in accident research, 3, 11-27.
Chang, F., Haque, M. M., Yasmin, S., & Huang, H. (2022). Crash injury severity analysis of E-Bike Riders: A random parameters generalized ordered probit model with heterogeneity in means. Safety science, 146, 105545.
Chang, F., Xu, P., Zhou, H., Chan, A. H., & Huang, H. (2019). Investigating injury severities of motorcycle riders: A two-step method integrating latent class cluster analysis and random parameters logit model. Accident Analysis & Prevention, 131, 316-326.
Chang, F., Yasmin, S., Huang, H., Chan, A. H., & Haque, M. M. (2022). Modeling endogeneity between motorcyclist injury severity and at-fault status by applying a Bayesian simultaneous random-parameters model with a recursive structure. Analytic methods in accident research, 36, 100245.
Chang, F., Yasmin, S., Huang, H., Chan, A. H. S., & Haque, M. M. (2021). Injury severity analysis of motorcycle crashes: A comparison of latent class clustering and latent segmentation based models with unobserved heterogeneity. Analytic methods in accident research, 32.
Chang, W. (2022). Taiwan’s ‘living hell’ traffic is a tourism problem, say critics.
Chen, C.-F., Fu, C., & Chen, Y.-C. (2023). Exploring tourist preference for Mobility-as-a-Service (MaaS) – A latent class choice approach. Transportation Research Part A: Policy and Practice, 174.
Chen, C.-F., Fu, C., & Siao, P.-Y. (2023). Exploring electric moped sharing preferences with integrated choice and latent variable approach. Transportation Research Part D: Transport and Environment, 121, 103837.
Chen, F., Song, M., & Ma, X. (2019). Investigation on the injury severity of drivers in rear-end collisions between cars using a random parameters bivariate ordered probit model. International journal of environmental research and public health, 16(14), 2632.
Chiou, Y.-C., & Fu, C. (2013). Modeling crash frequency and severity using multinomial-generalized Poisson model with error components. Accident Analysis & Prevention, 50, 73-82.
Chiou, Y. C., Fu, C., & Ke, C.-Y. (2020). Modelling two-vehicle crash severity by generalized estimating equations. Accident Analysis & Prevention, 148, 105841.
Chiou, Y. C., Hwang, C. C., Chang, C. C., & Fu, C. (2013). Modeling two-vehicle crash severity by a bivariate generalized ordered probit approach. Accid Anal Prev, 51, 175-184.
Chu, T. D., Miwa, T., Bui, T. A., Nguyen, Q. P., & Vu, Q. H. (2022). Examining unobserved factors associated with red light running in Vietnam: A latent class model analysis. Transportation safety and environment, 4(1), tdac006.
de Lapparent, M. (2008). Willingness to use safety belt and levels of injury in car accidents. Accident Analysis & Prevention, 40(3), 1023-1032.
Eccarius, T., & Lu, C.-C. (2020). Powered two-wheelers for sustainable mobility: A review of consumer adoption of electric motorcycles. International Journal of Sustainable Transportation, 14(3), 215-231.
Eluru, N., Bagheri, M., Miranda-Moreno, L. F., & Fu, L. (2012). A latent class modeling approach for identifying vehicle driver injury severity factors at highway-railway crossings. Accid Anal Prev, 47, 119-127.
Eluru, N., Bhat, C. R., & Hensher, D. A. (2008). A mixed generalized ordered response model for examining pedestrian and bicyclist injury severity level in traffic crashes. Accid Anal Prev, 40(3), 1033-1054.
Eluru, N., & Yasmin, S. (2015). A note on generalized ordered outcome models. Analytic methods in accident research, 8, 1-6.
Esmaili, A., Aghabayk, K., & Shiwakoti, N. (2022). Latent Class Cluster Analysis and Mixed Logit Model to Investigate Pedestrian Crash Injury Severity. Sustainability, 15(1), 185.
Fang, Z., Yuan, R., & Xiang, Q. (2024). An exploratory investigation into the influence of risk factors on driver injury severity in angle crashes: A random parameter bivariate ordered probit model approach. Traffic injury prevention, 25(1), 70-77.
Fountas, G., & Anastasopoulos, P. C. (2017). A random thresholds random parameters hierarchical ordered probit analysis of highway accident injury-severities. Analytic methods in accident research, 15, 1-16.
Fountas, G., Anastasopoulos, P. C., & Mannering, F. L. (2018). Analysis of vehicle accident-injury severities: A comparison of segment-versus accident-based latent class ordered probit models with class-probability functions. Analytic methods in accident research, 18, 15-32.
Gaweesh, S. M., Ahmed, I., & Ahmed, M. M. (2023). Analysis Framework to Assess Crash Severity for Large Trucks on Rural Interstate Roads Utilizing the Latent Class and Random Parameter Model. Transportation Research Record, 03611981231158627.
Hsu, Y.-C., Shiu, Y.-M., Chou, P.-L., & Chen, Y.-M. J. (2015). Vehicle insurance and the risk of road traffic accidents. Transportation Research Part A: Policy and Practice, 74, 201-209.
Hua, C., Fan, W., Song, L., & Liu, S. (2023). Analyzing the injury severity in overturn crashes involving sport utility vehicles: latent class clustering and random parameter logit model. Journal of transportation engineering, Part A: Systems, 149(3), 04022153.
Huang, Y., & Meng, S. (2019). Automobile insurance classification ratemaking based on telematics driving data. Decision Support Systems, 127, 113156.
Kim, J.-K., Ulfarsson, G. F., Shankar, V. N., & Mannering, F. L. (2010). A note on modeling pedestrian-injury severity in motor-vehicle crashes with the mixed logit model. Accident Analysis & Prevention, 42(6), 1751-1758.
Kim, S. H. (2023). How heterogeneity has been examined in transportation safety analysis: A review of latent class modeling applications. Analytic methods in accident research, 100292.
Kim, S. H., & Mokhtarian, P. L. (2023). Finite mixture (or latent class) modeling in transportation: Trends, usage, potential, and future directions. Transportation Research Part B: Methodological, 172, 134-173.
Li, J., Fang, S., Guo, J., Fu, T., & Qiu, M. (2021). A Motorcyclist-Injury Severity Analysis: A Comparison of Single-, Two-, and Multi-Vehicle Crashes Using Latent Class Ordered Probit Model. Accid Anal Prev, 151, 105953.
Li, Y., & Fan, W. D. (2019). Modelling severity of pedestrian-injury in pedestrian-vehicle crashes with latent class clustering and partial proportional odds model: A case study of North Carolina. Accid Anal Prev, 131, 284-296.
Mannering, F. L., & Bhat, C. R. (2014). Analytic methods in accident research: Methodological frontier and future directions. Analytic methods in accident research, 1, 1-22.
Mannering, F. L., Shankar, V., & Bhat, C. R. (2016). Unobserved heterogeneity and the statistical analysis of highway accident data. Analytic methods in accident research, 11, 1-16.
MOTC. (2023). National Road Safety Information Inquiry Website - Total Annual Fatalities and 30-Day Fatality Count.
NPA. (2022). Police Statistical Report for Week 26 of the Year 2022. National Police Agency, Ministry of the Interior
Phuksuksakul, N., Yasmin, S., & Haque, M. M. (2023). A random parameters copula-based binary logit-generalized ordered logit model with parameterized dependency: application to active traveler injury severity analysis. Analytic methods in accident research, 38.
Ramaswamy, V., DeSarbo, W. S., Reibstein, D. J., & Robinson, W. T. (1993). An empirical pooling approach for estimating marketing mix elasticities with PIMS data. Marketing Science, 12(1), 103-124.
Razi-Ardakani, H., Mahmoudzadeh, A., Kermanshah, M., & Shukla, S. K. (2020). What factors results in having a severe crash? a closer look on distraction-related factors. Cogent Engineering, 6(1).
Russo, B. J., Savolainen, P. T., Schneider, W. H., & Anastasopoulos, P. C. (2014). Comparison of factors affecting injury severity in angle collisions by fault status using a random parameters bivariate ordered probit model. Analytic methods in accident research, 2, 21-29.
Russo, B. J., Yu, F., & Smaglik, E. J. (2023). Examination of factors associated with fault status and injury severity in intersection-related rear-end crashes: Application of binary and bivariate ordered probit models. Safety science, 164.
Salehian, A., Aghabayk, K., Seyfi, M., & Shiwakoti, N. (2023). Comparative analysis of pedestrian crash severity at United Kingdom rural road intersections and Non-Intersections using latent class clustering and ordered probit model. Accident Analysis & Prevention, 192, 107231.
Savolainen, P. T., Mannering, F. L., Lord, D., & Quddus, M. A. (2011). The statistical analysis of highway crash-injury severities: A review and assessment of methodological alternatives. Accident Analysis & Prevention, 43(5), 1666-1676.
Schneider, W. H. t., Savolainen, P. T., Van Boxel, D., & Beverley, R. (2012). Examination of factors determining fault in two-vehicle motorcycle crashes. Accid Anal Prev, 45, 669-676.
Sfeir, G., Abou-Zeid, M., Rodrigues, F., Pereira, F. C., & Kaysi, I. (2021). Latent class choice model with a flexible class membership component: A mixture model approach. Journal of choice modelling, 41.
Sfeir, G., Rodrigues, F., & Abou-Zeid, M. (2022). Gaussian process latent class choice models. Transportation Research Part C: Emerging Technologies, 136, 103552.
Shaheed, M. S., & Gkritza, K. (2014). A latent class analysis of single-vehicle motorcycle crash severity outcomes. Analytic methods in accident research, 2, 30-38.
Shannon, D., Murphy, F., Mullins, M., & Eggert, J. (2018). Applying crash data to injury claims-an investigation of determinant factors in severe motor vehicle accidents. Accident Analysis & Prevention, 113, 244-256.
Sobhani, A., Eluru, N., & Faghih-Imani, A. (2013). A latent segmentation based multiple discrete continuous extreme value model. Transportation Research Part B: Methodological, 58, 154-169.
Song, D., Yang, X., Yang, Y., Cui, P., & Zhu, G. (2023). Bivariate joint analysis of injury severity of drivers in truck-car crashes accommodating multilayer unobserved heterogeneity. Accident Analysis & Prevention, 190, 107175.
Song, L., Fan, W. D., Li, Y., & Wu, P. (2021). Exploring pedestrian injury severities at pedestrian-vehicle crash hotspots with an annual upward trend: A spatiotemporal analysis with latent class random parameter approach. Journal of safety Research, 76, 184-196.
Srinivasan, K. K. (2002). Injury severity analysis with variable and correlated thresholds: ordered mixed logit formulation. Transportation Research Record, 1784(1), 132-141.
Sun, M., Sun, X., & Shan, D. (2019). Pedestrian crash analysis with latent class clustering method. Accident Analysis & Prevention, 124, 50-57.
Sun, Z., Xing, Y., Wang, J., Gu, X., Lu, H., & Chen, Y. (2022). Exploring injury severity of bicycle-motor vehicle crashes: A two-stage approach integrating latent class analysis and random parameter logit model. Journal of Transportation Safety & Security, 14(11), 1838-1864.
Wali, B., Ahmed, A., Iqbal, S., & Hussain, A. (2017). Effectiveness of enforcement levels of speed limit and drink driving laws and associated factors – Exploratory empirical analysis using a bivariate ordered probit model. Journal of Traffic and Transportation Engineering (English Edition), 4(3), 272-279.
Wang, S., Li, F., Wang, Z., & Wang, J. (2022). A random parameter bivariate probit model for injury severities of riders and pillion passengers in motorcycle crashes. Journal of Transportation Safety & Security, 14(8), 1289-1306.
Washington, S., Karlaftis, M. G., Mannering, F., & Anastasopoulos, P. (2020). Statistical and econometric methods for transportation data analysis. CRC press.
Weiss, A. A. (1993). A bivariate ordered probit model with truncation: Helmet use and motorcycle injuries. Journal of the Royal Statistical Society: Series C (Applied Statistics), 42(3), 487-499.
Wen, C.-H., Huang, W.-W., Fu, C., & Chou, P.-Y. (2013). A latent class generalised nested logit model and its application to modelling carrier choice with market segmentation. Transportmetrica A: Transport Science, 9(8), 675-694.
Xiao, D., Šarić, Ž., Xu, X., & Yuan, Q. (2022). Investigating injury severity of pedestrian–vehicle crashes by integrating latent class cluster analysis and unbalanced panel mixed ordered probit model. Journal of Transportation Safety & Security, 15(2), 83-102.
Xin, C., Guo, R., Wang, Z., Lu, Q., & Lin, P.-S. (2017). The effects of neighborhood characteristics and the built environment on pedestrian injury severity: A random parameters generalized ordered probability model with heterogeneity in means and variances. Analytic methods in accident research, 16, 117-132.
Yamamoto, T., & Shankar, V. N. (2004). Bivariate ordered-response probit model of driver’s and passenger’s injury severities in collisions with fixed objects. Accident Analysis & Prevention, 36(5), 869-876.
Yasmin, S., & Eluru, N. (2013). Evaluating alternate discrete outcome frameworks for modeling crash injury severity. Accident Analysis & Prevention, 59, 506-521.
Yasmin, S., Eluru, N., Bhat, C. R., & Tay, R. (2014). A latent segmentation based generalized ordered logit model to examine factors influencing driver injury severity. Analytic methods in accident research, 1, 23-38.
Yasmin, S., Eluru, N., & Pinjari, A. R. (2015). Analyzing the continuum of fatal crashes: A generalized ordered approach. Analytic methods in accident research, 7, 1-15.
Zhang, Z., Li, H., Hu, H., & Ren, G. (2022). How yielding cameras affect consecutive pedestrian-vehicle conflicts at non-signalized crosswalks? A mixed bivariate generalized ordered approach. Accid Anal Prev, 178, 106851.
Zhang, Z., Li, H., & Ren, G. (2023). Investigating jaywalker crossing risks from the sequential-conflict perspective: A grouped random parameters generalized ordered probit model. Accid Anal Prev, 189, 107145.
Zhou, J., Zheng, T., Dong, S., Mao, X., & Ma, C. (2022). Impact of Helmet-Wearing Policy on E-Bike Safety Riding Behavior: A Bivariate Ordered Probit Analysis in Ningbo, China. Int J Environ Res Public Health, 19(5).
Zou, W., Wang, X., & Zhang, D. (2017). Truck crash severity in New York city: an investigation of the spatial and the time of day effects. Accident Analysis & Prevention, 99, 249-261.